Inconsistent fueling. Injector heat soak?

[realspeeddan]

I’m having trouble with a CTS-V running vastly different fuel trims in the first 5 to 10 minutes after startup compared to when the car is more heat soaked from longer running time.

It is a 2011 CTS-V with a comp cam, ID850 injectors, long tube headers, upper and lower SC pulleys.

I did the complete tune using the same method as I used on several previous CTS-Vs, mapping the MAF separately and them mapping the VE equations using the BlueCat tool. I’m using an NTK wideband and generally follow the tuning procedure outline by Greg Banish (eficalibrator) from the classes I have taken.

I thought I had the car dialed in very well, but a few days later the customer came back with a lean code. Sure enough, when I checked, the fuel trims were 18% lean at idle and low rpm.

So I re-mapped the SD section and rechecked the MAF, made a few adjustments until the fuel trims were perfect and sent the car on its way.

A week later the car is back. The fuel trims are now 22% lean at idle with coolant almost up to full temperature. But after the car runs a little longer the fuel trims are pretty close to dead on, and just above idle the car is too rich and sometimes stumbles, as if I need to undo the enrichment I did when the car came back last week.

I thought we might have a vacuum leak that “sealed up” as the motor got hotter, so we smoke tested the intake and exhaust systems with the car cold and found no leaks.

I thought the factory o2 sensors might just be slow to warm up with the long tubes, so I reinstalled the wideband and checked. The wideband matches the factory O2 switching points pretty well, so I’ve ruled that out for now.

Right now I’m stuck. If anyone could take a look at the tunes and datalogs and offer suggestions I would appreciate it.

Dan

-------------------------
The wideband was going through the warm-up cycle during the first 18 seconds of the datalog, so pay no attention to the “AFR error inverse” or “Wideband O2” during the first 18 seconds.

Yeah, I know that the fuel trims are off at higher rpms when I free revved it neutral. Yeah, I know that The MAF g/s doesn’t perfectly match the VE g/sec. But both of these things actually matched better in an earlier version of the tune before I started trying to fix the temperature inconsistency. I’ll straighten that out once I get the car to be consistent.

Attachment 35477

[Mike@G-Force]

You know the V's run a few different fuel pressures.

Normal 60.2
high 65.3
low 36.3
cold start 60.2
default 65.3
fail 34.8

[IDRIVEAG8GT]

OP, you smashing your head on the wall?

[5_Liter_Eater]

Could be a tiny exhaust leak upstream of, or even at the O2 bung(s). Are both banks lean?

[realspeeddan]

Both banks are lean. Sometimes one side is leaner thant the other by about 8%, but that comes and goes. Most of the time both sides follow the same trend.

[realspeeddan]

OP, you smashing your head on the wall?

Just about ready to start doing that.

[eficalibrator]

What does your indicated air charge temp look like compared to raw measured IAT? Can you sneak a thermocouple into the plenum between the IC and ports?

It is possible that your heat soak characteristics are different enough from stock that your modeled air charge temp (which is biased by ECT) is not tracking reality with the change in pulley ratios and camshaft. Can you just make it follow IAT2 100% and see if that helps? If so, that would confirm which direction you need to go.

[realspeeddan]

Can you just make it follow IAT2 100% and see if that helps?

Any suggestion on how to do this? Unplug the upstream IAT sensor? Or change a setting in the tune?

[realspeeddan]

OP, you smashing your head on the wall?

Something tells me you've had similar frustrations and an opinion that you are keeping to yourself.

[realspeeddan]

Does any of the IVT gains get used in the base fuel calculation whenthe car is closed loop? HP tuners labels these as being used in open loop. Is it possible they are also applied in closed loop operation for the inital fuel injection calculation before O2 feeback is applied?


The same question also applies to the Open Loop InjTemp Gains.

[DSteck]

Make sure you're not trying to dial in the MAF during evap purge.

[realspeeddan]

The suggestion to look at EVAP purge as a factor got me looking, and I found this:
LTFT Purge cells: Reduction factor:
HP tuners definition:
“LTFT Purge Reduction Factor: his factor is used to reduce the effect of EVAP purge on fueling at higher airflows. Smaller numbers reduce the effects of purge on fueling.”

OK. The Question:
Does this table tell the ECU how much the fact that EVAP Purge is occurring WILL AFFECT the measured fuel trims? Or does it tell the ECU HOW MUCH PURGE TO ALLOW? or something different?

In other words, does Lowering the numbers in the table tell the ECU to: a) purge less, b) purge the same but compensate less, c) purge the same but compensate more, d) something totally different.

[realspeeddan]

OK. Part of the mystery solved. After a restart while tuning the car is going into "HOT FUEL" mode. While in Hot Fuel mode the fuel trims are significantly more negative than they are when the fuel pressure is in normal mode. So the question is, should I (or even can I) do anything to compensate for this, other than making sure the fuel pressure is in one of the normal modes while I'm re-mapping the calibration?


I am also going to unplug and and "fail" the purge solenoid circut and block off the evap line while I re-map the car, unless someone has a better way to make sure that evap purging isn't skewing my a/f reading while I'm mapping the VE and MAF.

[IDRIVEAG8GT]

Something tells me you've had similar frustrations and an opinion that you are keeping to yourself.

Well, yes I do. Simply try a Write Entire first and see if your problem goes away.

[IDRIVEAG8GT]

Does any of the IVT gains get used in the base fuel calculation whenthe car is closed loop? HP tuners labels these as being used in open loop. Is it possible they are also applied in closed loop operation for the inital fuel injection calculation before O2 feeback is applied?


The same question also applies to the Open Loop InjTemp Gains.

Yes, to answer your question, for the first second of cranking.

[realspeeddan]

Well, yes I do. Simply try a Write Entire first and see if your problem goes away.

Tried that 3 days ago.

[VodeAn]

Was this ever solved?? Having a similar issue.

[VodeAn]

This thread touches on this
https://www.hptuners.com/forum/showt...e-flashing-ECM

After reading that tread it seems the solution may be to just ignore the data until 5-10 minutes after a reflash??

[GHuggins]

You need to dial in your injector tip temperature. Whole reason this takes place especially on blower motors.